fosdem.org

Greg Kroah-Hartman will give a talk about writing and submitting your first Linux kernel patch at FOSDEM 2010.

Could you briefly introduce yourself?

Hi, I'm Greg, one of the many thousands of Linux kernel developers.

What will your talk be about, exactly?

How to write and submit your first Linux kernel patch. Isn't that the title? How can I get any more descriptive? :)

What do you hope to accomplish by giving this talk? What do you expect?

I hope every participant will know the exact steps that are needed to be able to write and submit a Linux kernel patch. Homework will be given out, so I expect everyone to actually submit a patch based on the talk. It would be nice to get a few hundred kernel patches from the attendees accepted into the next Linux kernel release.

What's so different about creating a Linux patch compared to creating a patch for other open source projects?

It is easier to submit and get a Linux kernel patch accepted as the Linux kernel developers try to make it as simple as possible.

How has the process of getting a patch accepted in the Linux kernel changed since the beginning of the Linux kernel? Which part of the process could still use some improvement?

The procedure has been documented much better, and the workflow is much more public now. As to what could still be improved, I do not know, what part of the process do people still have problems with?

Let's say I can pinpoint an error in the kernel source and I know how to solve it. How much work does it take me to create a patch in the right way and how long will it take on average before it's accepted?

It should only take a few minutes to create and submit a patch in the correct format, once you know the steps involved. As for how long it will take to get accepted, it all depends on the type of bugfix, and what subsystem it is sent to. Some maintainers try to handle all submissions within a week, but some take a few weeks to get through their patch queue based on workloads and other external factors.

How much of the kernel contributions are done by people that are not doing it as part of their job?

20% of all kernel contributions are done by people doing this "on their own".

Recently you started with the "Ask a kernel developer" column for LWN. How were the reactions on it?

Very good to start with, people seem to have stopped sending me questions recently. That means that either:

• all questions were answered and no one has anymore.
• people are not aware of the column anymore.

I need to do some more announcements to let people know more about this resource (part 1, part 2) if it is to continue in the future. Care to help out with this?

What's the most important piece of advice you would give to people creating their first Linux kernel patch?

Do not be afraid. The kernel is just another program that is very easy to understand and send changes to. Break any change up into small, logical pieces, and send them to the maintainers of the subsystem affected and there should not be any problem getting your change accepted.

What was your first Linux kernel patch?

It was a change on how the USB subsystem handled strings in the USB device descriptors.

Have you enjoyed previous FOSDEM editions?

I have never been to FOSDEM, but I have heard wonderful things about it, which is why I am very excited to be there this year.

This interview is licensed under a Creative Commons Attribution 2.0 Belgium License.